arXiv subjects

Yongcheng Li

Publications and source records attributed to Yongcheng Li.

11 recordsLinked to original sources

arXiv2025

Comparison of FTN-NOFDM and PCS-OFDM for Long-Haul Coherent Optical Communications

Single-wavelength 400G coherent optical communications have become a critical solution to meet the explosive traffic demands. However, the single-carrier modulation using low-order modulation formats requires a broader wavelength division multiplexing grid and expands the occupied optical bandwidth. In this paper, we propose the faster-than-Nyquist non-orthogonal frequency division multiplexing (FTN-NOFDM) to improve the spectral efficiency for long-haul coherent optical communications. The subcarrier number is set to eight to enable low-complexity FTN-NOFDM signal generation using a pruned inverse fast Fourier transform and inter-carrier interference (ICI) cancellation. To deal with the conventional timing recovery (TR) failure, a frequency tone-based TR is proposed for FTN-NOFDM. A time-domain multiple-input multiple-output equalizer is designed to update the tap coefficients based on outputs of conventional iterative detection (ID). To further mitigate ICI, a low-density parity check-assisted ID is integrated into the conventional ID module. FTN-NOFDM, probabilistic constellation shaping (PCS)-OFDM, and quadrature phase shift keying-OFDM are experimentally compared in a 400G coherent optical communication system over 11 cascaded 125-GHz wavelength-selective switches (WSSs) and 2000 km transmission. Results show that the FTN-NOFDM exhibits comparable WSS filtering tolerance to PCS-OFDM and superior nonlinearity tolerance, while PCS-OFDM achieves the best bit error ratio performance.

cs.NI↗

arXiv2025

Toward Rich Video Human-Motion2D Generation

Generating realistic and controllable human motions, particularly those involving rich multi-character interactions, remains a significant challenge due to data scarcity and the complexities of modeling inter-personal dynamics. To address these limitations, we first introduce a new large-scale rich video human motion 2D dataset (Motion2D-Video-150K) comprising 150,000 video sequences. Motion2D-Video-150K features a balanced distribution of diverse single-character and, crucially, double-character interactive actions, each paired with detailed textual descriptions. Building upon this dataset, we propose a novel diffusion-based rich video human motion2D generation (RVHM2D) model. RVHM2D incorporates an enhanced textual conditioning mechanism utilizing either dual text encoders (CLIP-L/B) or T5-XXL with both global and local features. We devise a two-stage training strategy: the model is first trained with a standard diffusion objective, and then fine-tuned using reinforcement learning with an FID-based reward to further enhance motion realism and text alignment. Extensive experiments demonstrate that RVHM2D achieves leading performance on the Motion2D-Video-150K benchmark in generating both single and interactive double-character scenarios.

cs.CV↗

arXiv2024

VisionLLM-based Multimodal Fusion Network for Glottic Carcinoma Early Detection

The early detection of glottic carcinoma is critical for improving patient outcomes, as it enables timely intervention, preserves vocal function, and significantly reduces the risk of tumor progression and metastasis. However, the similarity in morphology between glottic carcinoma and vocal cord dysplasia results in suboptimal detection accuracy. To address this issue, we propose a vision large language model-based (VisionLLM-based) multimodal fusion network for glottic carcinoma detection, known as MMGC-Net. By integrating image and text modalities, multimodal models can capture complementary information, leading to more accurate and robust predictions. In this paper, we collect a private real glottic carcinoma dataset named SYSU1H from the First Affiliated Hospital of Sun Yat-sen University, with 5,799 image-text pairs. We leverage an image encoder and additional Q-Former to extract vision embeddings and the Large Language Model Meta AI (Llama3) to obtain text embeddings. These modalities are then integrated through a laryngeal feature fusion block, enabling a comprehensive integration of image and text features, thereby improving the glottic carcinoma identification performance. Extensive experiments on the SYSU1H dataset demonstrate that MMGC-Net can achieve state-of-the-art performance, which is superior to previous multimodal models.

cs.CV↗

arXiv2024

Stain-aware Domain Alignment for Imbalance Blood Cell Classification

Blood cell identification is critical for hematological analysis as it aids physicians in diagnosing various blood-related diseases. In real-world scenarios, blood cell image datasets often present the issues of domain shift and data imbalance, posing challenges for accurate blood cell identification. To address these issues, we propose a novel blood cell classification method termed SADA via stain-aware domain alignment. The primary objective of this work is to mine domain-invariant features in the presence of domain shifts and data imbalances. To accomplish this objective, we propose a stain-based augmentation approach and a local alignment constraint to learn domain-invariant features. Furthermore, we propose a domain-invariant supervised contrastive learning strategy to capture discriminative features. We decouple the training process into two stages of domain-invariant feature learning and classification training, alleviating the problem of data imbalance. Experiment results on four public blood cell datasets and a private real dataset collected from the Third Affiliated Hospital of Sun Yat-sen University demonstrate that SADA can achieve a new state-of-the-art baseline, which is superior to the existing cutting-edge methods with a big margin. The source code can be available at the URL (\url{https://github.com/AnoK3111/SADA}).

cs.CV↗

arXiv2024

Static and Dynamic Routing, Fiber, Modulation Format, and Spectrum Allocation in Hybrid ULL Fiber-SSMF Elastic Optical Networks

Traditional standard single-mode fibers (SSMF) are unable to satisfy the future long-distance and high-speed optical channel transmission requirement due to their relatively large signal losses. To address this issue, the ultra-low loss and large effective area (ULL) fibers are successfully manufactured and expected to deployed in the existing optical networks. For such ULL fiber deployment, network operators prefer adding ULL fibers to each link rather than replace existing SSMFs, resulting in a scenario where both of SSMF and ULL fiber coexist on the same link. In this paper, we investigated the routing, fiber, modulation format, and spectrum allocation (RFMSA) problem in the context of an elastic optical network (EON) where ULL fiber and SSMF coexisting on each link under both the static and dynamic traffic demands. We formulated this RFMSA problem as a node-arc based Mixed Integer Linear Programming (MILP) model and developed Spectrum Window Plane (SWP)-based heuristic algorithms based on different fiber selection strategies, including spectrum usage based (SU), optical signal-to-noise ratio (OSNR) aware, ULL fiber first (UFF), and random strategies. Simulation results show that in the static traffic demand situation, the RFMSA algorithm based on the OSNR-aware (OA) strategy exhibits optimal performance, attaining a performance similar to that of the MILP model regarding the maximum number of frequency slots (FSs) used in the entire network. Moreover, in the dynamic traffic demand scenario, the SU strategy remarkably surpasses the other strategies in terms of the lightpath blocking probability.

cs.NI↗

arXiv2024

Domain-invariant Representation Learning via Segment Anything Model for Blood Cell Classification

Accurate classification of blood cells is of vital significance in the diagnosis of hematological disorders. However, in real-world scenarios, domain shifts caused by the variability in laboratory procedures and settings, result in a rapid deterioration of the model's generalization performance. To address this issue, we propose a novel framework of domain-invariant representation learning (DoRL) via segment anything model (SAM) for blood cell classification. The DoRL comprises two main components: a LoRA-based SAM (LoRA-SAM) and a cross-domain autoencoder (CAE). The advantage of DoRL is that it can extract domain-invariant representations from various blood cell datasets in an unsupervised manner. Specifically, we first leverage the large-scale foundation model of SAM, fine-tuned with LoRA, to learn general image embeddings and segment blood cells. Additionally, we introduce CAE to learn domain-invariant representations across different-domain datasets while mitigating images' artifacts. To validate the effectiveness of domain-invariant representations, we employ five widely used machine learning classifiers to construct blood cell classification models. Experimental results on two public blood cell datasets and a private real dataset demonstrate that our proposed DoRL achieves a new state-of-the-art cross-domain performance, surpassing existing methods by a significant margin. The source code can be available at the URL (https://github.com/AnoK3111/DoRL).

cs.CV↗

arXiv2024

Towards Cross-Domain Single Blood Cell Image Classification via Large-Scale LoRA-based Segment Anything Model

Accurate classification of blood cells plays a vital role in hematological analysis as it aids physicians in diagnosing various medical conditions. In this study, we present a novel approach for classifying blood cell images known as BC-SAM. BC-SAM leverages the large-scale foundation model of Segment Anything Model (SAM) and incorporates a fine-tuning technique using LoRA, allowing it to extract general image embeddings from blood cell images. To enhance the applicability of BC-SAM across different blood cell image datasets, we introduce an unsupervised cross-domain autoencoder that focuses on learning intrinsic features while suppressing artifacts in the images. To assess the performance of BC-SAM, we employ four widely used machine learning classifiers (Random Forest, Support Vector Machine, Artificial Neural Network, and XGBoost) to construct blood cell classification models and compare them against existing state-of-the-art methods. Experimental results conducted on two publicly available blood cell datasets (Matek-19 and Acevedo-20) demonstrate that our proposed BC-SAM achieves a new state-of-the-art result, surpassing the baseline methods with a significant improvement. The source code of this paper is available at https://github.com/AnoK3111/BC-SAM.

cs.CV↗

arXiv2020

Electromyogram (EMG) Removal by Adding Sources of EMG (ERASE) -- A novel ICA-based algorithm for removing myoelectric artifacts from EEG -- Part 2

Extraction of the movement-related high-gamma (80 - 160 Hz) in electroencephalogram (EEG) from traumatic brain injury (TBI) patients who have had hemicraniectomies, remains challenging due to a confounding bandwidth overlap with surface electromyogram (EMG) artifacts related to facial and head movements. In part 1, we described an augmented independent component analysis (ICA) approach for removal of EMG artifacts from EEG, and referred to as EMG Reduction by Adding Sources of EMG (ERASE). Here, we tested ERASE on EEG recorded from six TBI patients with hemicraniectomies while they performed a thumb flexion task. ERASE removed a mean of 52 +/- 12% (mean +/- S.E.M) (maximum 73%) of EMG artifacts. In contrast, conventional ICA removed a mean of 27 +/- 19\% (mean +/- S.E.M) of EMG artifacts from EEG. In particular, high-gamma synchronization was significantly improved in the contralateral hand motor cortex area within the hemicraniectomy site after ERASE was applied. We computed fractal dimension (FD) of EEG high-gamma on each channel. We found relative FD of high-gamma over hemicraniectomy after applying ERASE were strongly correlated to the amplitude of finger flexion force. Results showed that significant correlation coefficients across the electrodes related to thumb flexion averaged 0.76, while the coefficients across the homologous electrodes in non-hemicraniectomy areas were nearly 0. Across all subjects, an average of 83% of electrodes significantly correlated with force was located in the hemicraniectomy areas after applying ERASE. After conventional ICA, only 19% of electrodes with significant correlations were located in the hemicraniectomy. These results indicated that the new approach isolated electrophysiological features during finger motor activation while selectively removing confounding EMG artifacts.

eess.SP↗

arXiv2020

Electromyogram (EMG) Removal by Adding Sources of EMG (ERASE) -- A novel ICA-based algorithm for removing myoelectric artifacts from EEG -- Part 1

Electroencephalographic (EEG) recordings are often contaminated by electromyographic (EMG) artifacts, especially when recording during movement. Existing methods to remove EMG artifacts include independent component analysis (ICA), and other high-order statistical methods. However, these methods can not effectively remove most of EMG artifacts. Here, we proposed a modified ICA model for EMG artifacts removal in the EEG, which is called EMG Removal by Adding Sources of EMG (ERASE). In this new approach, additional channels of real EMG from neck and head muscles (reference artifacts) were added as inputs to ICA in order to "force" the most power from EMG artifacts into a few independent components (ICs). The ICs containing EMG artifacts (the "artifact ICs") were identified and rejected using an automated procedure. Simulation results showed ERASE removed EMG artifacts from EEG significantly more effectively than conventional ICA. Subsequently, EEG was collected from 8 healthy participants while they moved their hands to test the realistic efficacy of this approach. Results showed that ERASE successfully removed EMG artifacts (on average, about 75% of EMG artifacts were removed when using real EMGs as reference artifacts) while preserving the expected EEG features related to movement. We also tested the ERASE procedure using simulated EMGs as reference artifacts (about 63% of EMG artifacts removed). Compared to conventional ICA, ERASE removed on average 26% more EMG artifacts from EEG. These results indicate that using additional real or simulated EMG sources can increase the effectiveness of ICA in removing EMG artifacts from EEG. Combined with automated artifact IC rejection, ERASE also minimizes potential user bias.

eess.SP↗

arXiv2017

Robust Group LASSO Over Decentralized Networks

This paper considers the recovery of group sparse signals over a multi-agent network, where the measurements are subject to sparse errors. We first investigate the robust group LASSO model and its centralized algorithm based on the alternating direction method of multipliers (ADMM), which requires a central fusion center to compute a global row-support detector. To implement it in a decentralized network environment, we then adopt dynamic average consensus strategies that enable dynamic tracking of the global row-support detector. Numerical experiments demonstrate the effectiveness of the proposed algorithms.

cs.DC↗

arXiv2016

Multi-modal Fusion for Diabetes Mellitus and Impaired Glucose Regulation Detection

Effective and accurate diagnosis of Diabetes Mellitus (DM), as well as its early stage Impaired Glucose Regulation (IGR), has attracted much attention recently. Traditional Chinese Medicine (TCM) [3], [5] etc. has proved that tongue, face and sublingual diagnosis as a noninvasive method is a reasonable way for disease detection. However, most previous works only focus on a single modality (tongue, face or sublingual) for diagnosis, although different modalities may provide complementary information for the diagnosis of DM and IGR. In this paper, we propose a novel multi-modal classification method to discriminate between DM (or IGR) and healthy controls. Specially, the tongue, facial and sublingual images are first collected by using a non-invasive capture device. The color, texture and geometry features of these three types of images are then extracted, respectively. Finally, our so-called multi-modal similar and specific learning (MMSSL) approach is proposed to combine features of tongue, face and sublingual, which not only exploits the correlation but also extracts individual components among them. Experimental results on a dataset consisting of 192 Healthy, 198 DM and 114 IGR samples (all samples were obtained from Guangdong Provincial Hospital of Traditional Chinese Medicine) substantiate the effectiveness and superiority of our proposed method for the diagnosis of DM and IGR, compared to the case of using a single modality.

cs.CV↗